UNC Chapel Hill Opens Funding Round for Faculty and Staff AI Projects

For a lot of people inside a university, artificial intelligence still feels like something happening to them rather than something they get to shape. A new program at UNC Chapel Hill is trying to flip that script.

The AI Acceleration Program is now accepting applications from full-time faculty and staff across all schools and units. The goal is straightforward: fund projects that use AI to improve teaching, research, public service or the everyday operations that keep a large university running. Think of it less as a research grant for abstract experimentation and more like seed funding for tools people will actually use next semester, not five years from now.

That distinction matters. The program explicitly wants projects "scoped to deliver measurable impact and meaningful outcomes, not merely a proof of concept." In plain terms, a flashy demo isn't enough. Applicants need to show their idea will work for real students, researchers or staff within a year, and ideally have legs to grow beyond that.

Awards range from $5,000 to $30,000 in Azure computing credits, with up to $5,000 more available per project to pay student assistants. Those credits come with a deadline of their own: they must be spent before July 2027. The toolkit available to award winners includes Azure's cloud computing infrastructure, OpenAI's models, Copilot Studio and related Azure services, so teams aren't just getting money, they're getting access to the computing muscle that makes modern AI tools possible.

One scheduling note worth flagging: the application deadline has shifted. The original notice listed an earlier date, but the due date is now Oct. 20, 2026. Anyone who started a proposal under the old timeline should double check they're working from current information.

What Kinds of Projects Qualify

The program organizes its priorities around three of the university's core missions, and the examples it offers give a useful window into what organizers are hoping to see.

On the teaching and learning side, the focus is on tools that help students succeed, not just tools that showcase AI for its own sake. That could mean adaptive learning software that adjusts practice problems to an individual student's pace, similar to how a tutor might slow down or speed up based on how quickly someone grasps a concept. It could also mean systems that give students faster, more detailed feedback on writing or problem sets, or course redesigns that teach students how to use AI responsibly within their specific field of study. That last category matters more than it might first appear. As AI tools become embedded in nearly every profession, helping students use them critically, rather than blindly, is itself a form of workforce preparation.

On the research and scholarship side, the program is looking for projects that speed up discovery or open new analytical doors. Examples include machine learning approaches that accelerate analysis of scientific, clinical, environmental or social science data, and AI systems that help scholars find patterns or themes buried in archives, text, images, audio or collections that would take a human researcher far longer to sift through by hand. There's even room for creative work: the program cites AI-enabled systems that can enhance live performances by responding to audience reactions in real time, a reminder that this initiative isn't confined to labs and lecture halls.

The third category, operations, is where AI's more mundane but genuinely useful applications live. Virtual assistants that field routine questions, automation for repetitive administrative tasks, dashboards that help leaders make better decisions with real data. None of this is glamorous. But anyone who has waited on hold for a simple administrative question knows how much friction these tools could remove from daily university life.

Proposals will be judged against five criteria, and the weighting tells you a lot about what organizers actually value. Impact carries the most weight at 30 percent, measuring whether the project's expected outcomes are significant and whether there's a credible case it will benefit its intended users. Measurement plan and path to scale each account for 20 percent, essentially asking: how will you know if this worked, and can it grow beyond its initial pilot? Readiness and feasibility, also 20 percent, looks at whether the team is actually prepared to pull this off on a realistic timeline with the resources they're requesting. Proposal quality rounds things out at 10 percent, rewarding applications that are clear and well organized rather than padded.

Applicants need to complete a proposal template and submit it through a Qualtrics form by the Oct. 20, 2026 deadline. Reviews will happen throughout October, with award decisions expected by mid-November 2026. Projects can begin as soon as award notifications go out, so there's no long bureaucratic gap between funding and action.

Why It Matters

Universities have spent the last few years caught in a strange position: AI tools are reshaping how students learn, how research gets done and how institutions operate, often faster than any formal policy can keep pace with. Programs like this one represent an attempt to get ahead of that curve rather than simply react to it.

What's notable here isn't the technology itself. Azure credits and OpenAI access are increasingly common currency across higher education. What stands out is the emphasis on measurable, human-centered outcomes rather than innovation for its own sake. A dashboard that helps a department head make better staffing decisions, a feedback tool that saves a writing instructor hours of grading time, an adaptive learning system that catches a struggling student before they fall too far behind: these are modest-sounding wins. But they're the kind of wins that actually touch people's daily lives, and that's a more honest measure of AI's value than any headline about a breakthrough model.

For faculty and staff at Carolina weighing whether to apply, the message seems clear. This isn't about chasing the newest AI trend. It's about finding the places where a well-built tool, backed by real resources and a real plan to measure success, can make someone's work or someone's education meaningfully better.